© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-14B
    5. 20-RESID-BATCHTOPK-65K__L0-80
    7. 13029
    Prev
    Next
    INDEX
    Explanations

    fringe

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    adamkarvonen/qwen3-14b-saes/saes_Qwen_Qwen3-14B_batch_top_k/resid_post_layer_20
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    å·¥ä¸ļ大åѦ
    -0.21
    çļ¿
    -0.20
    ctors
    -0.17
    åħ´ä¸ļ
    -0.16
    arium
    -0.16
    name
    -0.16
    ãģĦãĤĭ
    -0.16
    xon
    -0.16
    stringstream
    -0.16
    lando
    -0.16
    POSITIVE LOGITS
    yat
    0.19
    NING
    0.19
    earch
    0.19
     warfare
    0.18
    ging
    0.18
    amble
    0.17
    ep
    0.17
     Warfare
    0.17
    ä¹ĭä½ľ
    0.16
    less
    0.16
    Activations Density 0.003%

    No Known Activations